1. Field of the Invention
This invention relates generally to the field of extracting usable energy from a moving fluid, more particularly to windmills.
2. Description of the Related Art
The basic design of windmills, whether for grinding grain, pumping water, or generating electricity, has not significantly changed in hundreds of years. A stationary vertical tower supports a single upwind horizontal-axis rotor, which may drive a load either directly, or, more usually, through a mechanical transmission. The traditional windmill tower is rigid, with many historical examples actually being made of stone. A single large rotor served well on these early machines, since a large rotor spins slowly with high torque, perfect for turning a stone to grind grain. The mass of such a large rotor, combined with the primitive state of technology of the day, precluded a serious consideration of a flexible tower.
Currently, the “single large rotor” design still prevails, despite the fact that today's electrical generators require a much higher rotational rate than yesterday's grindstone. Excessive bending deflection of the tower on these modern windmills is seen as sloppy, inefficient, and even dangerous. The axis of rotation of the rotor is perpendicular to the tower, so excessive bending of the tower would tend to reduce the incident angle of the wind on the disk of the rotor, reducing the effective swept area. With their hard mounting, the huge rotors and gargantuan machinery that supports them do not take kindly to being shaken about, due to stresses caused by inertial, vibrational, and coriolis type forces. The rigidity of the tower therefore protects the machinery from excessive wear or damage. Often, at the price of aesthetic clutter and reduced utility of the land below, guy wires are used to further stabilize the rigid tower. This basic prior art design has been slowly refined over the centuries, by improvements in tower construction, blade design, transmissions, materials science, control systems, etc. Current models, however, normally used for generating electricity, are still only barely feasible from an economic standpoint. The rigid, vertical tower is often the most expensive component of a wind turbine. Since wind velocity increases with height, and available power is proportional to the wind speed cubed, a taller tower will result in more power collected. Usually the rigid tower must be strong enough to support not only the huge rotor, but the driveshaft, generator, and associated gearbox as well, in addition to blade feathering mechanisms, yaw control apparatus for directional guidance, and associated electronics and auxiliary mechanisms, commonly weighing many tons. Access for maintenance personnel, such as an interior stairway or ladder, is often built-in. Erection and even maintenance of such an unwieldy wind energy conversion system often requires a crane and other expensive equipment, to lift the heavy machine components to and from the top of the tower. Deaths have resulted from accidents during these procedures.
The idea that the bending deflection which a tower is so naturally inclined to undergo could be embraced and utilized as advantageous, rather than avoided as a flaw, or minimized as an undesirable characteristic, has not yet found a place in modern windmill design. Despite a general feeling among many designers that there “must be a better way”, alternatives to the “standard model” have thus far proven not to be cost-effective. Aside from the vertical axis designs, such as those of Darrieus, nobody is yet successfully thinking “out of the box” so to speak. Designers have been as yet unable to break away from the traditional, basic, medieval design. As we begin a new millenium, the stationary, rigid windmill tower, a veritable relic of the stone age, with its azimuthally adjustable cap, having a geared mechanism with a horizontal driveshaft, supporting a single large upwind rotor, as developed in the middle ages, yet persists.
Once the decision is made to erect such an expensive rigid tower, it becomes important to maximize the size and efficiency of the rotor so as to justify this high cost. The decision to use a single large rotor, rather than many small rotors, is based on a desire for simplicity, and economy of scale, but results in a whole new series of expenses: First, the circular area subtended by a spinning rotor is proportional to the diameter squared, while the rotor's actual volume (and hence its mass), is proportional to the diameter cubed. In other words, the larger the rotor, the less wind it can capture in relation to its mass. The significance of this cannot be overemphasized: The amount of wind available per unit rotor mass is inversely proportional to the rotor diameter. This means that a 10-meter rotor will capture 100 times as much wind as a 1-meter rotor, but will weigh 1000 times as much! So as its diameter has increased by an order of magnitude, its subtended wind collecting area per unit mass has decreased by an order of magnitude.
Of course, 100 of these smaller rotors would each require individual physical support at an effective height, as well as either 100 individual generators, or a mechanical means to combine the rotation of the individual rotors. In the current state of the art, the increased complexity and consequent higher manufacturing and maintenance costs, as well as possible aesthetic clutter of such a multi-rotor technology, have weighed in favor of designs using a single large rotor, despite the disproportionately higher mass. The huge, multi-ton rotors employed must be carefully designed to maximize aerodynamic efficiency, with complex mechanisms both for feathering the blades and for orienting the rotor assembly (yaw control) according to prevailing wind conditions. Balance and resonances must be closely controlled to minimize harmful vibration. Means for protecting the massive rotor from self-destruction in high winds, such as a feathering mechanism, are normally required. Also, the momentum and relative rigidity of a large, upwind rotor make it slow to accelerate, so the extra energy available in momentary or localized gusts is not fully utilized.
A downwind design is well known to avoid several of these disadvantages. Since the downwind blades are pushed away from the tower, rather than toward it, they are unlikely to contact it. A downwind design can therefore use lighter, more flexible blades, which can simply bend to avoid damage in higher winds. These lighter, more flexible blades can also take better advantage of momentary gusts, due to their resilience and ease of acceleration. Finally, a downwind design requires no yaw control mechanism, as it will tend to naturally orient itself in the proper direction. In the current state of the art, however, despite all of these advantages, downwind designs are not favored, due to: A) the large wind-shadow of current state-of-the-art, thick, rigid, vertical towers. The wind-shadow reduces overall efficiency and can cause fatigue from stresses due to resonant vibrations induced by the repeated passage of the blades through the shadow. The upwind side of the tower is much less affected by wind-shadow. B) the fact that the generator is often horizontally mounted at the top of the tower, and the electricity must be somehow transported to the ground; Over time, with no active yaw control, simple electrical cables are likely to eventually become twisted too far in one direction, so that slip rings must be used to transmit the electric power with rotational freedom. Slip rings add complexity, and are not well-suited to larger installations. Once active yaw control is deemed necessary, the downwind design has lost its main advantage of passive yaw control, so an upwind design makes more sense.
Vertical-axis machines, such as a Darrius or a Savonius, also avoid many problems of single-rotor, upwind designs. For one, the aforementioned yaw control problem is nonexistent, since vertical axis turbines work equally well no matter what the direction of the wind. Also, the generator may be located at ground level, greatly reducing both the required tower strength and maintenance costs. While these advantages inherent in today's vertical-axis machines are certainly extremely desirable, they are outweighed by technical drawbacks.
The Darrieus type, for example, is not self-starting, and once started, does not collect as much energy, as smoothly, as an equivalent sized horizontal-axis rotor. Since its blades project upward it is perceived as “not needing a tower” and so is usually installed close to ground level. Such an installation may suffer from a large discrepancy in wind velocity between the tops and bottoms of its blades, since wind at ground level is markedly slowed by friction. And in actuality, of course, the tops of the blades must be supported by something, which normally turns out to be a rigid vertical tower of sorts, even if it turns with the blades. This tower, while not supporting a generator, must still be quite strong and substantially rigid to withstand the horizontal wind forces acting upon the blades without distorting. As with a horizontal axis rotor, the area subtended by a Darrieus type rotor is proportional to the diameter squared, while blade mass increases with the diameter cubed, so that larger rotors can collect less wind per unit mass. Again, available power per unit rotor mass is inversely proportional to rotor size. A heavier rotor needs a stronger tower and stronger bearings. Often, guy wires are used in an effort to satisfy this requirement for tower strength. Guy wires can require an additional bearing at the top of the rotor, produce visual clutter, and reduce the agricultural or other viability of the land below.
A Savonius turbine has some of the advantages of the Darrius, being omnidirectional, and is self-starting, but unfortunately a Savonius is very inefficient. Combining the two vertical-axis machines, with a Darrius stacked atop a Savonius, serves to elevate the Darrius, and this combination is self-starting. Unfortunately, even this improved combination still swings a large, slow rotor, requiring a transmission to drive a faster-spinning generator, and utilizes a rigid, vertical tower at its core, with its inherent high cost. The most efficient use of such an expensive, rigid tower, in the current state of the art, is still an upwind, horizontal axis machine.
For a given wind speed, the blade tip speed for any size rotor is about the same, hence, the angular rate of rotation is inversely proportional to rotor diameter. For a given amount of driveshaft power, torque is inversely proportional to rotation rate. Consequently a large rotor will turn a shaft at low rotational speed, but with high torque. This slow rotation rate and consequent high torque of such a large rotor mandate the use of heavy-duty driveshafts and ratio gearing mechanisms in order to transmit the power to a faster-rotating generator. Contemporary generators must turn many times faster than today's large rotors in order to efficiently generate power. The gearbox required to achieve this increased rotational rate represents about 20% of the cost of current systems.
The rotor, driveshaft, transmission, generator, and associated equipment must then, as a unit, be provided with means for powered yaw control, (directional rotation about the vertical axis) to maintain proper aim into the wind. The weight of all this heavy-duty hardware must be supported by the rigid tower, further adding to the strength required of the tower, and its consequent cost. The rigid, vertical towers in modern wind generation systems consume about another 20% of total system cost. As you can see, current windmill designs result in a self-reinforcing cascade of upwardly spiraling costs. This cascade of cost begins with the decision to utilize a rigid, vertical tower. It is interesting to note that for centuries, we have taken wood from trees that readily bend with the wind, to build windmills that don't. More than one engineer has been frustrated that their best attempts to harness the light and fleet wind result in clanking, complicated, multi-ton monstrosities that literally shake themselves apart. There is a strong feeling among researchers that there must be some easier, more simple and cost-effective way to harness wind energy, if only we could find it. The challenge to wind energy development for the new millenium is to meet the wind on its own terms using the stronger, and more flexible materials now available.